Comparison of Surveillance Versus Aortic Endografting for Small Aneurysm Repair (CAESAR): Results from a Randomised Trial

Background: Randomised trials have failed to demonstrate benefit from early surgical repair of small abdominal aortic aneurysm (AAA) compared with surveillance. This study aimed to compare results after endovascular aortic aneurysm repair (EVAR) or surveillance in AAA <5.5 cm. Methods: Patients (50-79 years) with AAA of 4.1-5.4 cm were randomly assigned, in a 1:1 ratio, to receive immediate EVAR or surveillance by ultrasound and computed tomography (CT) and repair only after a defined threshold (diameter >= 5.5 cm, enlargement >1 cm /year, symptoms) was achieved. The main end point was all-cause mortality. Recruitment is closed; results at a median follow-up of 32.4 months are here reported. Results: Between 2004 and 2008, 360 patients (early EVAR = 182; surveillance = 178) were enrolled. One perioperative death after EVAR and two late ruptures (both in the surveillance group) occurred. At 54 months, there was no significant difference in the main end-point rate [hazard ratio (HR) 0.76; 95% confidence interval (CI) 0.30-1.93; p = 0.6] with Kaplan-Meier estimates of all-cause mortality of 14.5% in the EVAR and 10.1% in the surveillance group. Aneurysm-related mortality, aneurysm rupture and major morbidity rates were similar. Kaplan-Meier estimates of aneurysms growth >= 5 mm at 36 months were 8.4% in the EVAR group and 67.5% in the surveillance group (HR 10.49; 95% CI 6.88-15.96; p < 0.01). For aneurysms under surveillance, the probability of delayed repair was 59.7% at 36 months (84.5% at 54 months). The probability of receiving open repair at 36 months for EVAR feasibility loss was 16.4%. Conclusion: Mortality and rupture rates in AAA <5.5 cm are low and no clear advantage was shown between early or delayed EVAR strategy. However, within 36 months, three out of every five small aneurysms under surveillance might grow to require repair and one out of every six might lose feasibility for EVAR. Surveillance is safe for small AAA if close supervision is applied. Long-term data are needed to confirm these results.